NOVA Food Classification: Understanding Ultra-Processed Foods

In recent years, the concept of "ultra-processed food" has moved from an obscure academic classification to a central concern in public health nutrition. Headlines warn that ultra-processed foods make up 60% of calories in the American diet and are linked to obesity, heart disease, cancer, and premature death. But what exactly makes a food "ultra-processed," and how rigorous is the science behind these claims?

The answer begins with NOVA — not an acronym, but a name (from the Portuguese word for "new") — a food classification system developed by Brazilian researchers that categorizes all foods into four groups based on the nature, extent, and purpose of the industrial processing they undergo. NOVA has become the dominant framework for studying the health effects of food processing, referenced by the United Nations, the World Health Organization, and dietary guidelines in multiple countries.

This article examines the NOVA system's origins, its four classification groups, the growing body of evidence linking ultra-processed food to adverse health outcomes, and the criticisms it has faced.

Origins of the NOVA Classification

NOVA was developed by Carlos Monteiro and colleagues at the Center for Epidemiological Studies in Health and Nutrition at the University of São Paulo, Brazil. The foundational framework was published in a series of papers beginning in 2009, with the most comprehensive description appearing in Monteiro et al. (2019) in Public Health Nutrition.

The central insight behind NOVA is that the traditional way of classifying foods — by nutrient content (high-fat, high-fiber, etc.) or by food group (grains, dairy, protein) — misses a critical dimension: the degree and purpose of industrial processing. Two foods can have similar macronutrient profiles but very different health effects depending on how they were manufactured. A baked potato and a bag of potato chips both come from potatoes, but they undergo radically different processing that alters their physical structure, chemical composition, and metabolic effects.

The Four NOVA Groups

Group 1: Unprocessed or Minimally Processed Foods

These are foods obtained directly from plants or animals that have undergone no processing, or only minimal processing that does not add substances to the original food. Minimal processing includes removal of inedible parts, drying, crushing, grinding, pasteurization, refrigeration, freezing, and vacuum packaging.

Examples: Fresh fruits and vegetables, grains, legumes, nuts, eggs, fresh meat, fish, milk (pasteurized), plain yogurt (no added sugar), herbs, spices, tea, coffee, water.

The defining characteristic is that these processes do not add salt, sugar, oils, fats, or other substances. They are designed to preserve natural foods and make them suitable for storage, safe to eat, or easier to prepare.

Group 2: Processed Culinary Ingredients

These are substances extracted and purified from Group 1 foods or from nature, used in kitchens to prepare, season, and cook Group 1 foods. They are rarely consumed on their own.

Examples: Oils (olive, coconut, butter), sugar (white, brown), honey, maple syrup, salt, flour, starch, vinegar.

Group 2 items are the building blocks of home cooking. Their role in NOVA is important: when combined with Group 1 foods in home-prepared meals, they produce dishes that are generally considered healthy. The problem arises when these ingredients are used by industry in formulations designed to create the hyper-palatable, shelf-stable products in Group 4.

Group 3: Processed Foods

These are relatively simple products made by adding Group 2 ingredients (salt, sugar, oil) to Group 1 foods. Processing methods include canning, bottling, non-alcoholic fermentation, and artisanal techniques.

Examples: Canned vegetables (with salt), canned fish (in oil), cheese, freshly made bread (from flour, water, salt, yeast), salted or smoked meats, fruits preserved in syrup, beer and wine.

Group 3 products typically contain two or three ingredients and are recognizable versions of the original food. A jar of salted peanuts is still recognizably peanuts. A loaf of sourdough bread is still recognizably bread. The processing serves primarily to extend shelf life or enhance sensory qualities through traditional methods.

Group 4: Ultra-Processed Foods (UPFs)

This is the category that has generated the most scientific and public attention. Ultra-processed foods are industrial formulations made mostly or entirely from substances derived from foods and additives, with little if any intact Group 1 food.

Defining characteristics of UPFs:

• Contain ingredients not typically found in home kitchens: high-fructose corn syrup, hydrogenated oils, protein isolates, modified starches, maltodextrin, invert sugar
• Include cosmetic additives designed to make the final product palatable: flavors, flavor enhancers, colors, emulsifiers, humectants, sequestrants, bulking agents
• Are designed to be hyper-palatable, convenient, and have long shelf life
• Are heavily marketed, often with health claims

A useful rule of thumb: if the ingredient list contains substances you would not find in a home kitchen, or if the product could not plausibly be made at home using standard cooking techniques, it is likely ultra-processed.

The Evidence: Ultra-Processed Foods and Health

The volume of epidemiological evidence linking ultra-processed food consumption to adverse health outcomes has grown rapidly since 2018. Several large prospective cohort studies have produced consistent findings.

Key Studies and Findings

The UK Biobank Analysis

The UK Biobank provides particularly strong evidence due to its scale (approximately 200,000 participants with dietary data) and comprehensive health outcome tracking through the UK National Health Service. Analyses of this cohort have shown dose-response relationships between UPF consumption and cardiovascular disease, type 2 diabetes, and certain cancers — with the associations persisting after adjustment for BMI, total caloric intake, socioeconomic status, smoking, alcohol, and physical activity.

Mechanisms: Why Might UPFs Be Harmful?

The associations between UPF consumption and poor health outcomes are consistent and robust, but the causal mechanisms remain debated. Several pathways have been proposed:

• Nutrient profile: UPFs tend to be high in added sugars, sodium, saturated fat, and calories, while being low in fiber, vitamins, and minerals. However, some studies find that the association between UPF intake and health outcomes persists even after adjusting for nutrient composition, suggesting processing itself may matter.
• Food matrix disruption: Industrial processing disrupts the physical structure of foods, potentially altering how quickly nutrients are absorbed. Whole grains are digested slowly; refined grain products cause rapid blood sugar spikes.
• Additives: Emulsifiers (like carboxymethylcellulose and polysorbate 80) have been shown in animal studies to alter gut microbiome composition and promote intestinal inflammation. Non-nutritive sweeteners may affect glucose tolerance through microbiome-mediated pathways.
• Hyper-palatability: UPFs are engineered to maximize consumption through precise combinations of sugar, fat, salt, and texture. This can override normal satiety signals, leading to overconsumption. A landmark 2019 NIH study by Kevin Hall found that participants ate approximately 500 more calories per day when given ultra-processed diets compared to unprocessed diets matched for available calories, macronutrients, sugar, sodium, and fiber.
• Displacement: Calories spent on UPFs displace calories from whole foods, reducing overall diet quality.

Criticisms and Limitations of NOVA

Despite its widespread adoption, NOVA has faced significant criticism from researchers, food industry representatives, and some nutrition scientists.

Classification Ambiguity

The boundary between Group 3 and Group 4 is sometimes unclear. Is commercially produced whole-wheat bread with a few additives (to extend shelf life) ultra-processed? What about flavored yogurt with a small amount of added sugar and pectin? Different researchers have classified the same foods differently, raising concerns about reproducibility.

Heterogeneity Within Group 4

The ultra-processed category lumps together a vast range of products — from sugar-sweetened sodas to fortified breakfast cereals to plant-based meat alternatives. Critics argue that not all ultra-processed foods are nutritionally equivalent, and blanket recommendations to avoid all Group 4 foods could inadvertently discourage consumption of some nutritionally adequate or beneficial products.

Confounding and Causation

Observational studies cannot establish causation. People who consume high levels of UPFs may differ from those who do not in ways that are difficult to measure and adjust for — including socioeconomic status, food access, education, stress, sleep, and overall health awareness. While the Kevin Hall controlled feeding study provides some causal evidence, it was small (20 participants) and short-term (2 weeks per diet).

Nutrient-Based vs. Processing-Based Classification

Some researchers argue that the health effects attributed to "processing" are better explained by the nutrient profile of UPFs — high sugar, high sodium, low fiber. If you control for these nutrient factors, does the association with processing level remain? The evidence is mixed. Some studies find that nutrient adjustment attenuates but does not eliminate the association, while others find near-complete attenuation.

Global Adoption and Policy Impact

Despite these debates, NOVA has influenced dietary guidelines in several countries. Brazil's dietary guidelines (2014) were the first to incorporate processing-based recommendations, advising citizens to "make natural or minimally processed foods the basis of your diet" and to "avoid ultra-processed foods." France, Israel, and several Latin American countries have since incorporated similar language.

The WHO has also referenced the NOVA framework in its guidance on healthy diets, and the Pan American Health Organization (PAHO) uses NOVA as a basis for its nutrient profiling model.

Relevance to Nutrition Tracking

For nutrition tracking apps, NOVA provides a valuable lens that complements traditional nutrient-based analysis. While calorie counting tells you how much you ate and nutrient profiling tells you what nutrients you consumed, processing-level classification tells you something about the quality and nature of the food itself.

KCALM incorporates processing-level awareness as part of its food quality scoring. When a user logs a food, the app can flag whether it falls into the ultra-processed category, helping users build awareness of processing levels in their diet — not as a rigid rule, but as one of several quality signals alongside nutrient density, dietary pattern adherence, and variety.

The practical takeaway from the NOVA research is not that all processing is bad — pasteurization, freezing, and fermentation are all forms of processing that can preserve or enhance nutritional value. Rather, it is that the specific industrial techniques and additive-heavy formulations that characterize Group 4 foods deserve scrutiny, and that building a diet around minimally processed whole foods remains the most evidence-supported path to long-term health.

Citations:

• Monteiro, C. A., Cannon, G., Levy, R. B., Moubarac, J.-C., Louzada, M. L. C., Rauber, F., Khandpur, N., Cediel, G., Neri, D., Martinez-Steele, E., Baraldi, L. G., & Jaime, P. C. (2019). Ultra-processed foods: what they are and how to identify them. Public Health Nutrition, 22(5), 936-941.
• Hall, K. D., Ayuketah, A., Brychta, R., Cai, H., Cassimatis, T., Chen, K. Y., ... & Zhou, M. (2019). Ultra-Processed Diets Cause Excess Calorie Intake and Weight Gain: An Inpatient Randomized Controlled Trial of Ad Libitum Food Intake. Cell Metabolism, 30(1), 67-77.
• Lane, M. M., Gamage, E., Du, S., Ashtree, D. N., McGuinness, A. J., Gauci, S., ... & Marx, W. (2024). Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses. BMJ, 384, e077310.
• Fiolet, T., Srour, B., Sellem, L., Kesse-Guyot, E., Allès, B., Méjean, C., ... & Touvier, M. (2018). Consumption of ultra-processed foods and cancer risk: results from NutriNet-Santé prospective cohort. BMJ, 360, k322.
• Ministry of Health of Brazil. (2014). Dietary Guidelines for the Brazilian Population. 2nd Edition. Brasília: Ministry of Health.